The discriminatory power of ordinal measures - towards a new coefficient

Perspective distortion, occlusion and specular reflection are challenging problems in shape-from-stereo. In this paper we review one recently published area-based stereo matching algorithm (Bhat and Nayar, 1998) designed to be robust in these cases. Although the algorithm is an important contribution to stereo-matching, we show that its coefficient has a low discriminatory power, which leads to a significant number of multiple best matches. In order to cope with this drawback we introduce a new normalized ordinal correlation coefficient. Experiments showing the behavior of the proposed coefficient are performed on various datasets including real data with ground truth. The new coefficient reduces the occurrence of multiple best matches to almost zero per cent. It also shows a more robust and equally accurate behavior. These benefits are achieved at almost no additional computational costs.

[1]  Shree K. Nayar,et al.  Ordinal Measures for Image Correspondence , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Shree K. Nayar,et al.  Ordinal measures for visual correspondence , 1996, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[3]  M. Kendall.  Rank Correlation Methods , 1949 .

[4]  Gérard G. Medioni,et al.  3-D Surface Description from Binocular Stereo , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Otmar Kolednik,et al.  Reconstruction of surface topographies by scanning electron microscopy for application in fracture research , 1996 .

[6]  Martin A. Fischler,et al.  Computational Stereo , 1982, CSUR.

[7]  Robyn A. Owens,et al.  2D feature detection via local energy , 1997, Image Vis. Comput..

[8]  Stefan Scherer,et al.  Robust adaptive window matching by homogeneity constraint and integration of descriptions , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[9]  Ramin Zabih,et al.  Non-parametric Local Transforms for Computing Visual Correspondence , 1994, ECCV.

[10]  Dinkar N. Bhat.  An evolutionary measure for image matching , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[11]  Carlo Tomasi,et al.  Depth Discontinuities by Pixel-to-Pixel Stereo , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).